HAL Id: hal-02951189https://hal.umontpellier.fr/hal-02951189

Submitted on 28 Sep 2020

HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, estdestinée au dépôt et à la diffusion de documentsscientifiques de niveau recherche, publiés ou non,émanant des établissements d’enseignement et derecherche français ou étrangers, des laboratoirespublics ou privés.

First case of brucellosis caused by an amphibian-typeBrucella

Nicolas Rouzic, Ludovic Desmier, Marie-Estelle Cariou, Eugénie Gay, JeffreyFoster, Charles Williamson, François Schmitt, Mikael Le Henaff, Alain Le

Coz, Aurélien Lorléac’h, et al.

To cite this version:Nicolas Rouzic, Ludovic Desmier, Marie-Estelle Cariou, Eugénie Gay, Jeffrey Foster, et al.. First caseof brucellosis caused by an amphibian-type Brucella. Clinical Infectious Diseases, Oxford UniversityPress (OUP), In press, pp.ciaa1082. �10.1093/cid/ciaa1082�. �hal-02951189�

 1/12  

First case of brucellosis caused by an amphibian-type Brucella

Running title : 1st infection by frog Brucella in human

Nicolas Rouzic1 (MD), Ludovic Desmier2,3, Marie-Estelle Cariou4 (PharmD), Eugénie Gay4

(MD), Jeffrey T. Foster5 (PhD), Charles H.D. Williamson5 (PhD), François Schmitt4

(PharmD, PhD), Mikael Le Henaff6 (MD), Alain Le Coz6 (MD) 1 (MD),

Jean-Philippe Lavigne2,3 (MD, PhD) 2,3 (PhD) and Anne Keriel2,3* (PhD)

1 Unité de Médecine Interne Maladies Infectieuses, Groupe Hospitalier Bretagne Sud,

Lorient, France

2 Centre National de Référence (CNR) des Brucel la, CHU de Nîmes, Nîmes, France

3 VBMI, U1047, INSERM, Université de Montpellier, Nîmes, France

4 Laboratoire de biologie médicale, Groupe Hospitalier Bretagne Sud, Lorient, France

5 The Pathogen & Microbiome Institute (PMI), Northern Arizona University, Arizona, USA

6 Service de Pneumologie, Groupe Hospitalier Bretagne Sud, Lorient, France

Keywords : brucellosis, Brucella, amphibian, atypical, pulmonary infection

Corresponding author: Anne Keriel, anne.keriel@inserm.fr, Phone: +33 4 66 02 81 57

Alternate correspon david.o-callaghan@umontpellier.fr,

Phone: +33 4 66 02 81 46

 2/12  

 3/12  

A BST R A C T : We report the first case of brucellosis caused by an isolate whose genome is

identical that of a frog isolate from Texas, demonstrating the zoonotic potential of

amphibian-type Brucella. Importantly, with such atypical Brucella, correct diagnosis cannot

be performed using routine serological tests or identification methods.

 4/12  

Brucellosis, an infectious disease caused by members of the Brucella genus, remains

one of the most common bacterial zoonotic diseases worldwide and is endemic in most

developing countries, mainly in areas with extensive farming [1]. Brucella are Gram-negative

facultative intracellular bacteria that can infect many organs and soft tissues. Brucella

melitensis (goats) is the predominant species causing human brucellosis, but B. abortus

(cattle) or B. suis (swine) may also be involved. Transmission is commonly due to the

consumption of unpasteurized dairy products or direct exposure to diseased animals.

Metropolitan France is officially free from bovine brucellosis since 2005 and autochthonous

human infections are extremely rare. Human brucellosis is a chronic debilitating infection

with non-specific clinical presentation, often presenting as a flu-like syndrome. Symptoms

occur after an incubation period of 2 weeks, or up to several months in some cases.

Mid-January 2019 (day 0), a French 28-years-old patient was hospitalized in the

general hospital of Lorient (France) for an exploration of polyadenopathies associated with

multiple pulmonary condensations.

Four months earlier, the patient had reported a left cervical lymphadenopathy,

associated with general degradation in condition and weight loss (8 kg in 3 months). A

cervical ultrasound performed on day -30 confirmed fluid accumulation in the left cervical

region and revealed two others in the maxillary region. The patient also reported intense night

sweats (without known fever) and a loose cough that resolved spontaneously. The patient,

also affected by Wilson's disease, previously suffered from hypertension requiring multiple

ligations of varicose veins. His treatment at day 0 was: propranolol, pyridoxine, D-

penicillamine, paroxetine and hydroxyzine.

On admission (day 0), the patient had altered general status, moderate dyspnea without

cough and night fever (39°C). Clinical examination confirmed a left cervical

lymphadenopathy. Heart and lungs auscultations were normal. Initial laboratory tests (day 1)

showed a C-reactive protein level of 37 mg/L, indicative of an active infection, but total

leukocyte count of within the normal range (5×109/L with 60% polymorphonuclear cells.

 5/12  

HIV serology was negative. Blood cultures were negative after five days incubation. A

cervico-thoracic and abdominal CT scan revealed multiple mediastinal adenopathies,

pulmonary parenchymal condensations, cystic emphysematous lesions in the right upper lobe

and splenomegaly (Figure S1).

On day 6, a left cervical lymph node was surgically removed, which triggered fever in

the patient. Histopathologic examination of the excised tissue revealed an aspecific epithelio-

gigantocellulary granuloma with caseous necrosis, without acid-fast bacilli (later confirmed

by negative mycobacterial culture). Aerobic cultures of the tissue were positive after 48h

incubation on Chocolate + polyvitex, Drigalski and Columbia + 5% sheep blood agar plates

(bioMérieux). Colonies were about 1 mm in diameter, circular, glistening, smooth, non-

pigmented and non-haemolytic. Culture of endobronchial aspiration from the patient in a

Mycobacteria Growth Indicator Tube (Becton Dickinson) amplified bacteria that, upon sub-

culturing on a Columbia plate, gave colonies with similar appearance. Bacteria were Gram-

negative coccobacilli and showed positive reactions for catalase and cytochrome-oxidase, but

negative result for indole production. The bacterial isolate was analyzed by MALDI-TOF MS

using a Biotyper (Bruker Daltonics) with the version 7 (V7) of the IVD (In Vitro Diagnostics)

database. The identification result was Ochrobactrum anthropi (score 1.49). Because of the

low score and inconsistency with phenotypic observations, the isolate was re-analyzed using

Bruker's Security-Relevant Library database and the identification result was Brucella

melitensis (score 2.23). A treatment with oral doxycycline (200 mg once daily) and

rifampicin (15 mg/Kg/day) was thus initiated and the isolate sent to the national reference

centre (CNR Brucella) for confirmation of identification and further analysis. MALDI-TOF

MS re-analysis using a Vitek MS (bioMérieux) with the IVD database V3.2 confirmed that

the isolate belongs to the Brucella genus. However, the use of a recently developed, more

complete, RUO (Research Use Only) database [2] rather suggested that it was a B. inopinata

(score 99.9%). A blood sample drawn on day 1 was also submitted to several Brucella

serological tests: Rose Bengal (Brucelloslide-Test, bioMérieux), Wright agglutination

 6/12  

(Brucella Wright, Bio-Rad), ELISA (Virclia IgM or IgM, Vircell) and immunocapture test

(BrucellaCapt, Vircell). All tests were negative.

After 6 weeks of antibiotic therapy, the patient reported resolution of his malaise and

fatigue but presented a left cervical abscess under a surgical scar. Culture of an aspirate from

this abscess was negative, but Brucella DNA was detected by PCR amplification of the

Brucella-specific sequence IS711 [3]. The persistence of deep-seated lymphadenopathy

associated with multiple pulmonary condensations was confirmed by a CT scan showing an

unsatisfactory progression. The antibiotic treatment was thus prolonged for a total of 3

months, to which the patient responded favorably, remaining clear of symptoms 6 months

after the end of treatment.

An epidemiological survey was conducted to try identifying the origin of the infection.

The patient declared no travel abroad in the past five years and no consumption of

unpasteurized dairy products. However, as part of his professional activity (animal keeper

specializing in exotic animals), he reported many direct contacts with reptiles, amphibians,

rodents or birds, several imported from abroad.

The bacterial isolate was characterized using a Bruce-Ladder multiplex PCR [4] on

genomic DNA and phylogenic positioning based on MALDI-TOF MS spectra. This revealed

that this isolate is most similar to the B. inopinata-like isolates BO2 and B13-0095, isolated

from a patient with severe pulmonary infection [5] and from a frog [6], respectively (Figures

S2 and S3). The isolate, thereafter named BO3 (for Brucella inopinata-like 3), showed

susceptibility to all tested antibiotics, as for B. melitensis and the other B. inopinata(-like)

strains (Table S1). Several features however clearly distinguished BO3 from classical

Brucella spp. First, slide agglutination using a polyclonal serum anti-Brucella (ANSES,

Maisons-Alfort, France) was negative. We also found that BO3 is highly motile, like other B.

inopinata(-like) strains [6], and flagellated (Figure S4), which is completely novel for this

bacterial genus.

Sequencing of BO3 genome revealed two chromosomes of 2,181,202 bp and 1,197,108

 7/12  

bp, respectively, with a GC content of 57.11%, similar to other Brucella species. No plasmid

was detected. A genome-based phylogenetic tree confirmed that BO3 belongs to the group of

Brucella, together with the human isolates BO1 (USA, 2005 [7]) and BO2

(Australia, 2007 [5]) and a growing number of frog isolates [8], and is extremely closely

related to the frog isolate B13-0095 [6] (Figure 1). Annotation of the BO3 genome using

PATRIC (https://www.patricbrc.org) identified the virB operon (encoding a type IV secretion

system) on chromosome 2. As with strains BO2 and B13-0095, the wbk operon (encoding the

perosamine-based O-antigen in classical Brucella species) is absent from BO3 genome and,

instead, a gene cluster encoding a system for L-rhamnose uptake and catabolism is present,

suggesting an atypical lipopolysaccharide. BO3 DNA also contains a cluster of genes

involved in ectoine catabolism (Figure S5), a genetic feature specific to amphibian-type

Brucella [6,8]. When calculating genome-to-genome distance, the DNA-DNA hybridization

(DDH) estimate between strains BO3 and B13-0095 was 99.9%, indicating that these

genomes are identical. As a comparison, the DDH estimate between BO3 and BO2 was

89.8%, 84.9% with BO1, 83.3% with 09RB8471 (African bullfrog [9]), 83.2% with

141012304 (bluespotted ribbontail ray [10]), 81.5% with 10RB9215 (African bullfrog) and

81.4% with 09RB8910 (African bullfrog). Altogether, our analyses indicate that this new

strain BO3 is identical to the B. inopinata-like strain B13-0095 that was isolated from a Pac-

Man frog in Texas in 2010 [6].

This is thus the first report of human brucellosis caused by an amphibian-type Brucella.

The scarcity of reports may be due to the difficulty to identify them as Brucella spp. Because

of their rapid growth rate on plates (< 24h) and because phenotypic identification tests (such

as API 20NE) misidentify them as Ochrobactrum spp. with a high confidence level [6,11].

Moreover, since some strains have an LPS that is chemically and antigenically different from

that of classical Brucella, infection cannot be detected using commonly available serological

tests and the isolates cannot be identified using specific typing serum. Finally, when using

MALDI-TOF MS as an identification method, it should be noted that the Biotyper (Bruker),

 8/12  

with its Security Relevant database, and the Vitek MS (bioMérieux), with its most recent

IVD database , would identify such atypical isolates as B. mel itensis or Brucella

spp., respectively. Correct identification as B. inopinata(-like) is currently only possible using

the latest version RUO database Saramis on a Vitek MS [2].

The animal reservoir(s) of B. inopinata(-like) strains remain to be identified. Isolation

from different animals (either wild caught or kept in zoological collections, pet store or by

private breeder [8]) on several continents suggests that these bacteria are widespread. Recent

isolation from a ray also suggests that all vertebrates (including fish and reptiles) may be

suitable hosts. We could not detect traces of Brucella in cultures of samples taken from the

, suggesting that exposure occurred during his past professional

activity as an exotic animal keeper (notably, the patient declared being in contact with Pac-

Man frogs). We speculate that B. inopinata(-like) bacteria may be part of the normal flora of

some animals and become opportunistic pathogens for their host upon unfavorable conditions

such as stress caused by transport (importation of wild-caught animals), changes in breeding

facilities (i.e. quarantine protocols) or contact with other species creating opportunities for

cross-species infection. Also, pathogenicity for humans may be linked to lowered

immunocompetency.

Medical practitioners and diagnostic laboratories should be aware of the existence of

these pathogens. Prevention measures should be taken, especially for exotic animal keepers

and veterinarians, who should limit direct contact with potentially infected animals.

 9/12  

A C K N O W L E D G M E N TS

This work was supported by the Institut National de la Santé et de la Recherche Médicale

(Inserm), Université de Montpellier and Santé Publique France. We are grateful to Chantal

Cazevieille (Electron Microscopy Platform, Montpellier RIO Imaging) for the visualisation

of the BO3 cells. We are also very grateful to MVDr Paredes (Clinique Vétérinaire Copernic,

to Fanny Broussard for the analysis of

the animals samples.

C O MPL I A N C E W I T H E T H I C A L ST A ND A RDS

Conflict of interest: All authors do not declare any conflict of interest.

Ethical approval: Publication of case reports does not require approval by an Institutional

Review Board.

Informed consent: The patient gave his written informed consent for the publication of this

case report.

Nucleotide sequence accession numbers and deposition

The genome sequences of isolate BO3 are available from the GenBank database under

accession numbers CP047232-CP047233 (BioProject PRJNA597040).

 10/12  

R E F E R E N C ES

1.

brucellosis in developing countries: Moving away from improvisation. Comp Immunol

Microbiol Infect Dis 2013;

2. Mesureur J, Arend S, Cellière B, et al. A MALDI-TOF MS database with broad genus

coverage for species-level identification of Brucella. PLoS Negl Trop Dis 2018;

3. Sanjuan-Jimenez R, Morata P, Bermúdez P, Bravo MJ, Colmenero JD. Comparative

Clinical Study of Different Multiplex Real Time PCR Strategies for the Simultaneous

Differential Diagnosis between Extrapulmonary Tuberculosis and Focal Complications

of Brucellosis. PLoS Negl Trop Dis 2013;

4. López-Goñi I, García-Yoldi D, Marín CM, et al. Evaluation of a multiplex PCR assay

(Bruce-ladder) for molecular typing of all Brucella species, including the vaccine

strains. J Clin Microbiol 2008;

5. Tiller R V, Gee JE, Lonsway DR, et al. Identification of an unusual Brucella strain

(BO2) from a lung biopsy in a 52 year-old patient with chronic destructive pneumonia.

BMC Microbiol 2010; 10:23.

6. Soler-Lloréns PF, Quance CR, Lawhon SD, et al. A Brucella spp. Isolate from a Pac-

Man Frog (Ceratophrys ornata) Reveals Characteristics Departing from Classical

Brucellae. Front Cell Infect Microbiol 2016; 6:116.

7. De BK, Stauffer L, Koylass MS, et al. Novel Brucella Strain (BO1) Associated with a

Prosthetic Breast Implant Infection. J Clin Microbiol 2008; 46:43 49.

8. Brucella in

amphibians: are we facing novel emerging pathogens? J Appl Microbiol 2016;

9. Al Dahouk S, Köhler S, Occhialini A, et al. Brucella spp. of amphibians comprise

genomically diverse motile strains competent for replication in macrophages and

survival in mammalian hosts. Sci Rep 2017; 7:44420.

10. Eisenberg T, Ri??e K, Schauerte N, Geiger C, Blom J, Scholz HC. Isolation of a novel

 11/12  

atypical Brucella strain from a bluespotted ribbontail ray (Taeniura lymma). Antonie

van Leeuwenhoek, Int J Gen Mol Microbiol 2016; :1 14.

11. Tiller R V, Gee JE, Frace MA, et al. Characterization of novel Brucella strains

originating from wild native rodent species in North Queensland, Australia. Appl

Environ Microbiol 2010; 76:5837 5845.

12. Scholz HC, Revilla-Fernández S, Dahouk S Al, et al. Brucella vulpis sp. Nov., isolated

from mandibular lymph nodes of red foxes (vulpes vulpes). Int J Syst Evol Microbiol

2016;

 12/12  

F I G UR E L E G E ND

F igure 1: The amphibian-type isolate B O3 belongs to the emerging group of atypical

Brucella. BO3 genomic DNA was extracted (DNeasy UltraClean Microbial Kit, Qiagen) and

sequenced using MinION (Oxford Nanopore Technologies) and Illumina MiSeq (Illumina)

platforms. Sequence reads were de novo assembled with Unicycler v0.4.7, processed with

Circlator v1.5.5 and polished with Pilon v1.23) resulting in a high quality genome. A

genome-based tree was then constructed using the reference strains of all Brucella species

NUCmer; core genome single nucleotide polymorphisms were called within NASP; and a

maximum likelihood phylogeny was inferred with IQ-TREE. This phylogenetic tree shows

Brucella, together with B. inopinata BO1

[7], B. inopinata(-like) BO2 [5], and isolates from Australian rodents (NF2653 and 83/13

[11]), red foxes (F60 and F965 [12]), a bluespotted ribbontail ray [10] and frogs (B13-0095,

09RB8910, 10RB9215 and 09RB8471 [6,9]).

Rouzic N . et al., manuscr ipt for C linical Infectious Diseases

SUPPL E M E N T A R Y M A T E RI A L

F igure S1: Initial CT scan showing hilar and subcarinal adenopathies (A) and parenchymal

condensations (B).

F igure S2: Identification of the patient bacterial isolates as Brucella. inopinata-like. Bruce-Ladder

multiplex PCR was performed on two bacterial isolates BRSO-2019-086 and BRSO-2019-089, recovered

after puncture of two different lymph nodes in the patient at day 6 and -7, respectively. Bruce-Ladder

profiles were compared to classical (B. melitensis, B. abortus, B. suis and B. canis) or atypical (B.

inopinata strain BO1 or B. inopinata-like strains BO2 and frog B13-0095) Brucella isolates after agarose

gel electrophoresis. Left = molecular-weight size marker (size in bp). Ø = negative control.

F igure S3: Proximity of M A L DI-T O F MS spectra. Dendrogram constructed from MALDI-TOF MS

spectra using SARAMIS (bioMérieux). Similarity between spectra is quantified by the proportion of

common masses; the dotted line (65%) indicating a threshold above which bacteria are considered as

significantly different. This representation shows the distance between Brucella strains belonging to the

classical or to the atypical clade of this genus. This highlights clustering of the two patient isolates

BRSO-2019-086 and BRSO-2019-089 (arrows) Brucella, and confirms the

clonality of these two isolates, which were thereafter considered as a single strain named BO3 (for

Brucella inopinata-like 3).

Table S1: Antimicrobial susceptibility tests by disk diffusion assay. Bacterial isolates were plated on Mueller-Hinton agar with 5% sheep blood

(bioMérieux), as established by the Clinical Laboratory Standards Institute and considering suggestions for Gram-negative bacteria. The cut-offs (in

brackets) and the measured inhibition diameters are indicated in mm. 16M = B. melitensis biovar 1 reference strain, BO1 = B. inopinata clinical isolate [1],

BO2 = B. inopinata-like clinical isolate [2], BO3 = B. inopinata-like clinical isolate (this case), B13-0095 = B. inopinata-like frog isolate [3].

Rifampicin

Tetracycline

Gentamycin

Amikacin

Trimethoprim +

Sulfamethoxazol

Ciprofloxacin

Ofloxacin

Amoxicillin

Amoxicillin +

Clavulanic

acid

16M 40 45 32 32 42 40 32 52 49

B O1 32 43 23 22 28 35 32 40 40

B O2 32 50 24 24 20/25 24/24 24/23 50 42

B O3 30 40 22 22 16/25 30/30 30/26 40 32

B13-

0095 32 40 22 22 15/18 30/29 24/23 40 36

F igure S4: Brucella B O3 isolate is flagellated and motile. A) Swimming assays on soft-agar plates

showing that BO3 is motile. As a control, strains BO1 and BO2 are also motile while B. melitensis strain

16M (classical Brucella) is not, in agreement with previous observations [3]. B) Visualization of BO3 by

with a polar flagellum (scale bar = 500 nm).

F igure S5: Detection of the E ctoin genes cluster by PC R. A) Specific PCR strategy: two independent

reactions were performed -actggctcttcctcaagac- -

accgacaacatcgcaatc- -gcaagttcttcgacccatc- to amplify either a 370 bp fragment

common to all Brucella species (as a positive control) or a 1234 bp fragment when the bacterial genomic

DNA contains the Ectoin genes cluster, respectively. B) -

2019-086 and BRSO-2019-089) compared to other Brucella isolates (B. melitensis 16M, B. inopinata

strain BO1, B. inopinata-like strains BO2 and frog B13-0095) after agarose gel electrophoresis. Left =

molecular-weight size marker (size in bp). Ø = negative controls.

R E F E R E N C ES

1. De BK, Stauffer L, Koylass MS, et al. Novel Brucella Strain (BO1) Associated with a Prosthetic

Breast Implant Infection. J Clin Microbiol 2008; 46:43 49.

2. Tiller R V, Gee JE, Lonsway DR, et al. Identification of an unusual Brucella strain (BO2) from a

lung biopsy in a 52 year-old patient with chronic destructive pneumonia. BMC Microbiol 2010;

10:23.

3. Soler-Lloréns PF, Quance CR, Lawhon SD, et al. A Brucella spp. isolate from a Pac-Man frog

(Ceratophrys ornata) reveals characteristics departing from classical brucellae. Front Cell Infect

Microbiol 2016;